DESCRIPTION (provided by investigator): Background and goals: Each year in the US, up to 490,000 pregnant women undergo elective induction, an intervention to bring on labor without a medical reason. Whether this causes risks to the woman or fetus is unclear. Only one recent clinical trial studied elective induction before 41 weeks of fetal gestation, and it found no difference in outcomes compared to letting the pregnancy continue (expectant management). Observational studies have found more cesarean delivery after induction, but most compared induction to spontaneous labor at the same gestational age, which is not the genuine clinical alternative. Instead, elective induction should be compared to the true alternative, expectant management. Most studies were too small to study uncommon but important outcomes, and none looked at whether risks differ by the mother's race/ethnicity or obesity. Our goal is to compare outcomes for elective induction at 38, 39 or 40 weeks' gestation versus expectant management, to help pregnant women and their providers make more informed decisions. Methods: A multiphase retrospective study of births from 2001 to 2011 will be conducted in two integrated health care systems. In Phase 1, the eligible population (206,000 singleton births) will be identified and classified by their apparent induction and outcome status based on electronic health plan and birth certificate data. In Phase 2, we will conduct 8000 brief medical record reviews to determine true exposure status (elective induction vs. expectant management). In Phase 3, we will review 3500 records in depth for true outcome status and risk factors. Phases 2 and 3 will oversample women with elective induction or rare outcomes, improving our power to address these specific aims: Aim 1: Compare the risk of the following outcomes after elective induction at 38, 39 or 40 weeks' gestation versus expectant management: cesarean delivery, postpartum hemorrhage, chorioamnionitis, neonatal intensive care unit stay, shoulder dystocia, low Apgar score, stillbirth, and neonatal death; and Aim 2: Examine how risks vary by the mother's race/ethnicity, parity, and pre-pregnancy obesity. Secondary analyses will develop an algorithm to more accurately identify elective inductions from electronic data. Impact: This research will compare elective labor induction at 38, 39 or 40 weeks' gestation to expectant management to provide new information about its risks or benefits for the mother and fetus. Our results could immediately affect clinical practice by helping women and providers make more informed choices together. They will also provide a solid foundation for practice guidelines and policy making.